1. Field of the Invention
The present invention relates to lasers, and particularly to manufacturing lasers using a monolithic platform that has a silicon base.
2. Description of Related Art
One important type of laser is a diode-pumped solid-state laser ("DPSSL"), which includes a solid-state gain medium pumped by optical radiation from a laser diode. Diode-pumped solid-state lasers can be useful in a wide variety of applications including laser display systems, optical data storage systems, medical instrumentation, and telecommunications. For small to moderate levels of optical power, one particularly useful diode-pumped solid-state laser is a microlaser, which comprises a short element (i.e. less than about five mm) of solid-state gain medium positioned in a resonant cavity. The pump beam is supplied by a semiconductor diode laser, which pumps the solid-state gain medium to provide energy to support laser operation within the resonant cavity. Two opposing reflective surfaces define the resonant cavity within which the gain medium is disposed. The opposing reflective surfaces may be formed directly on opposing ends of the solid-state gain medium, and the resulting structure is compact and reliable.
Optically-pumped solid-state lasers have many advantages relative to other lasers, including high efficiency, small size, and low cost when compared with other lasers. However, manufacturing difficulties have prevented high volume production and accordingly have prevented the cost reductions necessary to open up new markets and uses, as well as to make current uses more attractive from a price standpoint. If manufacturing processes could be improved to provide high volume and low cost, lasers could be utilized in numerous additional products and applications.
Lasers, including DPSSL's, have been assembled on optical benches to form laser assemblies. The optical benches or bases are the platform to which the components of the laser are attached. These bases have been formed by hand machining metals such as aluminum or copper. Such metals provide excellent heat dissipation characteristics, which are important for high power (e.g. greater than 100 milliwatt) operation. For example, copper has a thermal conductivity of about 300 W/m-.degree. C. The benches are typically coupled to thermoelectric ("TE") coolers which provide temperature control. However, the large coefficient of thermal expansion for these metals renders them unsuitable for precision placement of optical components thereon, because even a small relative movement between optical components can drastically degrade performance of the laser and optical processing functions. Alternative materials such as the alloy invar have been used which alleviate thermal expansion problems. However, these materials are generally poor thermal conductors, which lead to heat dissipation problems that can be expensive or impossible to overcome. Invar, for example, has a thermal expansion coefficient of about 1.5.times.10.sup.-6 m/.degree. C., and a thermal conductivity of about 16 W/m/.degree. C. It would be useful if an optical bench could be manufactured with a material that has a high thermal conductivity and a low coefficient of thermal expansion, and whose features can be formed with close tolerances in a low cost, reliable manner.
When metals such as copper and alloys such as invar are used as optical benches, each optical bench is machined individually, leading to inconsistencies between parts as well as adding significant cost. Furthermore, the inability of machining processes to achieve close tolerances requires substantial additional work (i.e. additional cost) to assemble a complete laser assembly. Invar particularly is very difficult to machine. It would be useful if batch processing methods such as precision etching of single-crystal silicon could be used to form bases identically and in large batches at low cost. However, three dimensional features in metals cannot be etched with sufficient precision, leaving machining as the only practical alternative for forming an optical bench of metals and alloys.
Lasers assemblies have been manufactured by hand by gluing free-standing optical components on optical benches in an exacting and laborious process For example, a skilled technician mounts each individual component on the optical benches, aligns each as it is glued into place, and then holds it in place for the length of time necessary to obtain a bond. Proper alignment of each component to within very narrow tolerances (e.g. five to ten microns) is critical because, without proper alignment, the laser will not operate. Due to the extremely narrow tolerances of optical systems, it can be difficult and time-consuming to achieve such tolerances with free-standing optical components. This can lead to excessive assembly time and reduced manufacturing yields, which translate directly to increased production costs.
Typically the laser diode is mounted on a first optical bench and the resonant cavity including the solid state gain medium is mounted on a separate second optical bench. Typically each of optical benches is provided with a separate TE cooler. One reason for the separate bases and TE coolers of conventional designs is to allow for thermal differences between the laser diode and the solid state gain medium or laser element (e.g. the diode and the laser element may require different operating temperatures for optimum performance). Of course this approach introduces additional difficulties in creating and maintaining proper alignment
Even if the laser is aligned adequately during such conventional production, the glue drying process may introduce a small but significant optical misalignment during the glue drying time. Furthermore, subsequent use and heating during operation may cause slight misalignments, which can substantially degrade laser performance, or even halt it altogether.
Although such a hand assembly process of free-standing optical components may be suitable for low volume, high cost production, it is inappropriate for medium to large scale production. Furthermore, the nature of the hand assembly of free-standing optical components makes it difficult to achieve the economies of scale that will be necessary to reduce the cost per unit of the lasers.
Another problem with such an assembly process is caused by the adhesives used to glue the laser components to the platform. Adhesives within a laser can "outgas" certain airborne contaminants. In a hermetically-sealed laser, these outgassed contaminants become deposited upon the laser components, leading to significant contamination of critical surfaces such as the laser crystal face. Such a buildup can be avoided by eliminating the hermetically sealed enclosure and providing a constant flow of air; however, the air flow increases the likelihood of particulate contamination of the optical surfaces. It would be useful to provide an assembly method that reduces or eliminates adhesives so that a hermetically sealed enclosure can be used to provide a reliable defense against particulate contamination.